On JSON and REBOL(rebol.com)
rebol.com
On JSON and REBOL
http://www.rebol.com/cgi-bin/blog.r?view=0522
1 comments
JSON was specified|devised by Douglas Crockford (and not Brendan Eich) - http://en.wikipedia.org/wiki/JSON
Crockford was a friend of Carl Sassenrath (creator of Rebol) and cited Rebol as an influence in the design of JSON - http://www.rebol.com/article/0423.html
Crockford was a friend of Carl Sassenrath (creator of Rebol) and cited Rebol as an influence in the design of JSON - http://www.rebol.com/article/0423.html
What I'm saying was that JSON was already an integral part of Javascript even though it wasn't named JSON at the time. I.e. what JSON is today already existed in Javascript before Crockford turned it into a spec. Crockford took an existing Javascript construct and turned it into a standalone spec. So, since JSON already existed in Javascript (but wasn't named so), how could REBOL have been an influence on it?
Literal notation utilised by JSON was indeed a part of Javascript, however JSON was a subset of Javascript specifically selected for cross-language messaging; by necessity pruning the parts of literal Javascript unsuited for this purpose and in part conforming to identical literals in other languages (see http://www.json.org/fatfree.html). It's unclear how without the limits set by the JSON spec that literal Javascript would have become a popular exchange format.
It took Crockford's Rebol-influenced eye to spot the potential for a minimalist though somewhat expressive exchange format based on elementary block structures, and near-genius to carve it out of the spartan confines of the Javascript syntactic swamps. Sadly not being Rebol, JSON does not include words (see 'Reserved Words' http://javascript.crockford.com/survey.html), dates, urls, email addresses (see http://www.rebol.com/rebolsteps.html); and will forever be burdened with comma-delimiters.
It took Crockford's Rebol-influenced eye to spot the potential for a minimalist though somewhat expressive exchange format based on elementary block structures, and near-genius to carve it out of the spartan confines of the Javascript syntactic swamps. Sadly not being Rebol, JSON does not include words (see 'Reserved Words' http://javascript.crockford.com/survey.html), dates, urls, email addresses (see http://www.rebol.com/rebolsteps.html); and will forever be burdened with comma-delimiters.
JSON also does not permit comments, which makes it unsuitable for many applications such as handwritten configuration files. :-/ It would be tempting to use JSON for such things, but it requires too much markup...and without comments it's pretty dead in the water. Things like YAML have been succeeding in this space instead...much to the annoyance of Rebol advocates.
But there is some potential for copying the successful aspects of these formats, while bringing back that Rebol flair with REN (REadable Notation):
https://github.com/rebolek/REN
Just basically Rebol "taking back JSON" and competing with the likes of YAML. We'll have to wait and see where that goes.
But there is some potential for copying the successful aspects of these formats, while bringing back that Rebol flair with REN (REadable Notation):
https://github.com/rebolek/REN
Just basically Rebol "taking back JSON" and competing with the likes of YAML. We'll have to wait and see where that goes.
I am reminded of a Native American comedian who had a routine, in which he was describing what he would do if he got caught taking a car. "I didn't steal it, officer. I discovered it." :-)
If you listen to what Douglas says, he does say it's more a "discovery" than an "invention". It was a pattern he'd seen in a few places, and he realized the need for its standardization. In his view, Rebol was a strong influence on that mindset. Hence he credits it in that view, and he feels people should take a look at what the language has to offer.
Prior to the open sourcing in December I was an outspoken critic of Rebol's proprietary nature...but since my demands were met, I now feel comfortable advocating it. (And I do so on a semi-daily basis.) Please do feel free to come chat with us and RebolBot:
http://chat.stackoverflow.com/rooms/291/rebol-and-red
If you listen to what Douglas says, he does say it's more a "discovery" than an "invention". It was a pattern he'd seen in a few places, and he realized the need for its standardization. In his view, Rebol was a strong influence on that mindset. Hence he credits it in that view, and he feels people should take a look at what the language has to offer.
Prior to the open sourcing in December I was an outspoken critic of Rebol's proprietary nature...but since my demands were met, I now feel comfortable advocating it. (And I do so on a semi-daily basis.) Please do feel free to come chat with us and RebolBot:
http://chat.stackoverflow.com/rooms/291/rebol-and-red
What does that mean? That Brendan Eich modeled JavaScript Object Notation (which is an integral part of the JavaScript language) using REBOL as an example?